Dihedral angle based dimensionality reduction for protein structural comparison

Structural comparison of proteins is considered as one of the highly focused research areas in the field of bioinformatics. Structural similarity estimation techniques using singular geometric parameters derived from spatial coordinates of protein structural atoms have been reported previously, but it is also ascertained that a single geometric parameter based structural estimation can result in misconstrued classification and functional interpretation. In this paper we propose a novel geometric parameters based comparison protocol that uses previously unexplored pair of dihedral angles NC_aNC_a and NCNC for similarity search. An orthonormal transformation is employed on the two-dimensional distribution for feature extraction and selective feature-space is represented in an indexing schema later used for similarity calibration. The results demonstrate the success of this dimensionality reduction based similarity measure in performing a rapid and length-independent similarity analysis of the protein structures.